Want to read this magazine in your preferred language? You can translate the entire document using any modern AI tool (such as Gemini, ChatGPT, or Claude). Just follow these simple steps:
Facial recognition is not merely a social courtesy; it is a foundational skill that begins at the very earliest stages of human development. Neurotypical newborns as young as 30 minutes old demonstrate a clear preference for human faces over non-face objects, a behavior that sets the stage for a lifetime of social learning. For autistic individuals, however, this trajectory is often different from birth. Research into infant gaze patterns shows that babies who later receive an autism diagnosis spend significantly less time looking at people's faces and more time looking at objects.
This early discrepancy means that autistic children often miss out on the critical linguistic and social information that is naturally acquired through facial attention. Because of this, many struggle to recognize familiar people or remember their names, even if they see them daily.
Traditional psychological literature often describes these challenges as "processing" deficits, suggesting that the brain has difficulty integrating facial features into a whole.
However, from a behavior analytic perspective, the term "processing" is problematic because it describes internal, unobservable events that cannot be directly measured or targeted for intervention. Instead of focusing on "poor facial memory" or "holistic processing issues," practitioners should look at the observable environmental variables and behaviorsâspecifically, stimulus control and attention.
When a child fails to recognize a familiar person, it is often not a failure of memory, but a failure of attention; the child has not spent enough time looking at the face for its unique features to exert control over the response of saying that person's name. If the child is not looking at the face when a name is provided, the relationship between the face and the name is never established. This often leads to "stimulus control errors," where the child might respond to the wrong cues, such as a person's voice or general body shape, rather than their facial identity.
To address these deficits effectively, practitioners should follow a hierarchical framework that prioritizes foundational skills over complex social maneuvers. The model consists of:
Too often, practitioners jump straight to teaching emotional recognition using flashcards. This is a mistake because if a child is not naturally attending to faces in their environment, they will never see the emotions happening in real-time, rendering their flashcard skills useless in the natural world.
The core reason many autistic children do not look at faces is that faces and social stimuli may not inherently function as reinforcers for them. While neurotypical children are highly sensitive to social consequences, autistic children are often more drawn to objects. Therefore, the first goal of any program should be reinforcer conditioning: making the human face a "conditioned reinforcer".
It is essential to distinguish between instructional control (telling a child "look at me") and motivational control (the child looking because they want to). Forcing eye contact can cause anxiety and discomfort for autistic individuals. Instead, we should pair our faces with existing reinforcers so that the child naturally seeks out the face because it has acquired value. This is similar to how a neutral piece of paperâmoneyâbecomes a powerful reinforcer only after it is paired with the things it can buy.
Based on the work of Dr. Francesca degli Espinosa, the following steps can be used to build a child's natural motivation to attend to faces:
Once a child is attending to faces more frequently, the focus shifts to facial identityârecognizing that "Person A" is different from "Person B." Rather than starting with "Who is this?" (Tact Training), it is more effective to use match-to-sample procedures.
Teaching emotions should begin with naturalistic, dynamic stimuli like videos or GIFs, which more closely resemble real life than static cards.
Practitioners should also capitalize on natural learning opportunities in the environment. For instance, if a classmate is crying, the adult can direct the child's attention to the peer and discuss how they might be feeling.
Crucially, we must also teach children to identify their own internal emotional states. Providing the language for an emotion while the child is actually experiencing it (e.g., saying "You look sad" when the child is upset about homework) is far more effective than looking at pictures of themselves later, because the internal feeling is an essential variable in learning the label.
The ultimate goal is to move toward Domain 3: understanding the social meaning and "why" behind facial expressions. This involves inferencing and perspective-taking, such as realizing a friend is sad because they spilled their juice. However, these complex social skills will never naturally emerge if we skip the foundational work of conditioning faces as reinforcers and establishing holistic facial attention. By building these "neural bridges" early and correctly, we can help autistic children learn more naturally from their social environment.